Comprehensive scheduling method and device for problem of interruptability holder preemption

Abstract

The present invention finds and defines a problem which possibly exists in an interrupt remapping mechanism under a virtual symmetric multiprocessing environment, i.e., a problem of Interruptability Holder Preemption (IHP). This problem causes the interrupt remapping mechanism to fail and reduces I/O performance of virtual machines. In order to solve the IHP problem, the present invention provides a proactive VCPU comprehensive scheduling method based on interruptability holder information. This method is based on Kernel-Based Virtual Machines (KVMs) which are widely used at present and paravirtualization network models thereof. By globally controlling and analyzing a running state of an interruptability holder and simultaneously considering global scheduling fairness of a system, a VCPU comprehensive scheduling method is established, which can effectively eliminate the IHP problem and obviously improve the I/O performance of the virtual machines.

The invention claimed is: 
     
       1. A comprehensive scheduling method for a problem of interruptability holder preemption, wherein the method comprises the following steps:
 step 1: a virtual network device receiving a data packet; 
 step 2: selecting a virtual CPU from a plurality of virtual CPUs as a destination virtual CPU, and then injecting arrival of a virtual interrupt notification data packet; 
 step 3: analyzing information of the virtual interrupt and marking the destination virtual CPU as an interruptability holder; 
 step 4: the destination virtual CPU receiving the virtual interrupt, triggering a corresponding interrupt processing function, disabling interruptability of the virtual network device and entering a polling mode to read the data packet; 
 step 5: determining whether a virtual CPU which needs to be scheduled out of a physical CPU is marked as the interruptability holder, and if so, continuously scheduling the virtual CPU which needs to be scheduled out of the physical CPU to run on the physical CPU and recording the virtual CPU which needs to be scheduled out of the physical CPU as a favored virtual CPU; and 
 step 6: determining in a polling manner whether the favored virtual CPU has already enabled the interruptability of the virtual network device, and if so, immediately scheduling the favored virtual CPU out of the physical CPU. 
 
     
     
       2. The comprehensive scheduling method for the problem of interruptability holder preemption according to  claim 1 , wherein, in step 2, one virtual CPU is selected as the destination virtual CPU according to an interrupt remapping mechanism. 
     
     
       3. The comprehensive scheduling method for the problem of interruptability holder preemption according to  claim 1 , wherein, in step 5, when the virtual CPU which needs to he scheduled out of the physical CPU is not marked as the interruptability holder, the virtual CPU is immediately scheduled out of the physical CPU. 
     
     
       4. The comprehensive scheduling method for the problem of interruptability holder preemption according to  claim 1 , wherein, in step 6, whether the favored virtual CPU has already enabled the interruptability of the virtual network device is determined in a VM Exit processing function.